1 / 31

Practical Statistics for Particle Physicists Lecture 1

Practical Statistics for Particle Physicists Lecture 1. Harrison B. Prosper Florida State University European School of High-Energy Physics Parádfürdő , Hungary 5 – 18 June, 2013. Outline. Lecture 1 Descriptive Statistics Probability & Likelihood Lecture 2

lizina
Download Presentation

Practical Statistics for Particle Physicists Lecture 1

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Practical Statistics for Particle PhysicistsLecture 1 Harrison B. Prosper Florida State University European School of High-Energy Physics Parádfürdő, Hungary 5 – 18 June, 2013

  2. Outline • Lecture 1 • Descriptive Statistics • Probability & Likelihood • Lecture 2 • The Frequentist Approach • The Bayesian Approach • Lecture 3 • Analysis Example

  3. Descriptive Statistics

  4. Descriptive Statistics – 1 Definition: A statistic is any function of the data X. Given a sample X = x1, x2, … xN, it is often of interest to compute statistics such as the sample average and the sample variance In any analysis, it is good practice to study ensemble averages, denoted by < … >, of relevant statistics

  5. Descriptive Statistics – 2 Ensemble Average Mean Error Bias Variance Mean Square Error

  6. Descriptive Statistics – 3 Exercise 1: Show this The MSE is the most widely used measure of closeness of an ensemble of statistics {x} to the true value μ The root mean square (RMS) is

  7. Descriptive Statistics – 4 Consider the ensemble average of the sample variance

  8. Descriptive Statistics – 5 The ensemble average of the sample variance has a negative bias of –V / N Exercise 2: Show this

  9. Descriptive Statistics – Summary The sample average is an unbiased estimate of the ensemble average The sample variance is a biased estimate of the ensemble variance

  10. Probability

  11. Probability – 1 Basic Rules 1. P(A) ≥ 0 2. P(A) = 1 if A is true 3. P(A) = 0 if A is false Sum Rule 4. P(A+B) = P(A) + P(B) if AB is false * Product Rule 5. P(AB) = P(A|B) P(B) * *A+B = A or B, AB = A and B, A|B = A given that B is true

  12. B AB A Probability – 2 By definition, the conditional probability of A given B is P(A) is the probability of A without restriction. P(A|B) is the probability of A when we restrictto the conditions under which B is true.

  13. B AB A Probability – 3 From we deduce Bayes’ Theorem:

  14. Probability – 4 A and B are mutually exclusive if P(AB) = 0 A and B are exhaustive if P(A) + P(B) = 1 Theorem Exercise 3: Prove theorem

  15. ProbabilityBinomial & Poisson Distributions

  16. Binomial & Poisson Distributions – 1 A Bernoulli trial has two outcomes: S= success or F = failure. Example: Each collision between protons at the LHC is a Bernoulli trial in which something interesting happens (S) or does not (F).

  17. Binomial & Poisson Distributions – 2 Let pbe the probability of a success, which is assumed to be the same at each trial. Since S and F are exhaustive, the probability of a failure is 1 – p. For a given orderO ofn trails, the probability Pr(k,O|n) of exactlyk successes and n – k failures is

  18. Binomial & Poisson Distributions – 3 If the order O of successes and failures is irrelevant, we can eliminate the order from the problem integrating over all possible orders This yields the binomial distribution which is sometimes written as

  19. Binomial & Poisson Distributions – 3 We can prove that the mean number of successes a is a=pn. Suppose that the probability, p, of a success is very small, then, in the limit p→ 0 and n → ∞, such that a is constant, Binomial(k, n, p) → Poisson(k, a). The Poisson distribution is generally regarded as a good model for a counting experiment Exercise 4: Prove it Exercise 5: Show that Binomial(k, n, p) → Poisson(k, a)

  20. Common Distributions and Densities

  21. Probability – What is it Exactly? There are at least two interpretations of probability: • Degree of belief in, or plausibility of, a proposition Example: It will snow inGenevaon Friday • Relative frequency of outcomes in an infinite sequence of identically repeated trials Example: trials: proton-proton collisions at the LHC outcome: the creation of a Higgs boson

  22. Likelihood

  23. Likelihood – 1 The likelihood function is simply the probability, or probability density function (pdf), evaluated at the observed data. Example 1: Top quark discovery (D0, 1995) p(D| d) = Poisson(D |d) probability to get a count D p(17|d) = Poisson(17|d) likelihood of observation D = 17 Poisson(D|d) = exp(-d) dD / D!

  24. Likelihood – 2 Example 2: Multiple counts Di with a fixed total count N This is an example of a multi-binnedlikelihood

  25. Likelihood – 3 Example 3: Red shift and distance modulus measurements of N = 580 Type Ia supernovae This is an example of an un-binnedlikelihood

  26. Likelihood – 4 Example 4: Higgs to γγ The discovery of the neutral Higgs boson in the di-photon final state made use of an an un-binnedlikelihood, where x = di-photon masses m = mass of new particle w = width of resonance s = expected signal b = expected background fs = signal model fb = background model Exercise 6: Show that a binned multi-Poisson likelihood yields an un-binned likelihood of this form as the bin widths go to zero

  27. Likelihood – 5 Given the likelihood function we can answer questions such as: • How do I estimate a parameter? • How do I quantify its accuracy? • How do I test an hypothesis? • How do I quantify the significance of a result? Writing down the likelihood function requires: • Identifying all that is known, e.g., the observations • Identifying all that is unknown, e.g., the parameters • Constructing a probability model for both

  28. Likelihood – 6 Example: Top Quark Discovery (1995), D0 Results knowns: D = 17 events B = 3.8 ± 0.6 background events unknowns: b expected background count s expected signal count d = b + s expected event count Note: we are uncertain about unknowns, so 17 ± 4.1 is a statement about d, not about the observed count 17!

  29. Likelihood – 7 Probability: Likelihood: where B = Q / k δB = √Q / k

  30. Summary Statistic A statistic is any function of potential observations Probability Probability is an abstraction that must be interpreted Likelihood The likelihood is the probability (or probability density) of potential observations evaluated at the observed data

  31. Tutorials Location: http://www.hep.fsu.edu/~harry/ESHEP13 Download tutorials.tar.gz and unpack tar zxvftutorials.tar.gz Need: Recent version of Root linked with RooFit and TMVA

More Related